https://youtu.be/7MC8WV-6Z_E?feature=shared

Tldr: rtx 4080 is about 2x faster than the M3 pro. M3 Max results are unknown.

There were some results on an arstechina forum where people claimed the M3 Max was as fast as a 4070 desktop but verify the numbers for yourself!

https://arstechnica.com/civis/threads/the-m3-max-pro-performance-comparison-thread.1496840/page-2

“Max GPU: 3:26, this doesn’t seem right based on prior items in this thread does it? CPU was pegged at 1526%

Figured it out. Had to set GPU preferences to Metal, it was defaulted to none.

Classroom, 300 Samples Max GPU: 19.84 seconds (Optimized Kernels)

Junkshop, 400 Samples @2560 x 1440 (is Junkyard something different?) Max GPU: 50.59 seconds (Optimizing Kernels) Max GPU: 48.64 seconds (Optimized Kernels)

Ollama: 65.96 tokens/s 60.51 tokens/s 59.97 tokens/s 62.41 tokens/s”

  • F9-0021@alien.topB
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    Blender is very optimized for CUDA and Optix. Apple silicon could be matching Nvidia in After Effects, but then gets curbstomped in Blender due to not having CUDA or Optix.

  • fishkeeper9000@alien.topB
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    While I do like new tech and certainly Apple M line of cpu+gpu chips on a single SOC are powerful, it needs to be compared to the right technology. Comparing an SOC with a single heat pipe versus a desktop class GPU with 2000 grams of heat sink is just not a good comparison. They are built for different things.

    The consumer class 4000 series rtx is not a workstation class GPU. It is designed for intensive grunt work for gamers. Latency is optimized for gaming. So high-frequency clock speeds. The higher the better latency.

    Apple M3 is geared towards the professional content creator class of user. Who does not like to be bothered by heavy GPU and fan noise. It won’t have the best latency but it has good access to the tools to help get professional work done.

    Until Apple develop their own “GPU”with high frequency and heavy cooling requirements, they will just be dumbing down games to candy crush style avg latency gaming.

  • DevAnalyzeOperate@alien.topB
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    The 4080 + i9-13900hx laptop has more total memory, more total bandwidth, and more total transistors, with the only real downside being that system resources are split across two processors on a bandwidth constrained bus. If correctly optimised, there are almost no multi-threaded workloads where I would expect the m3 Pro to actually outperform a 4080 + i9-13900hx in terms of performance itself.

    I don’t even nessecarily see the m3 pro as being a serious workstation processor. I see it as being almost a thin and light processor for people with moderate workloads like non-fanciful 4k editing or normalish development work. The battery life you’re going to get from it compared to a system running two independent processors on less recent lithography is going to be a LOT better. The single threaded performance is a lot better. The performance is still good enough for a heavy task here and there.

    I get the compulsion to compare roughly price equivalent laptops to eachother, but I don’t even feel like these two machines aren’t even really optimised for the same use case, so comparing their workstation performance since the computers are about the same price and being like “looks like Apple still has catching up to do” is almost missing the point. I feel like it makes more sense to compare this to the Max, even though the price isn’t the same. I’d compare this laptop to a cheaper high end AMD APU laptop because those laptops are a lot closer to eachother philosophically.

    • Famous_Wolverine3203@alien.topOPB
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      I know. But this was the only head to head 4080 to M3 Max comparision i could find.

      I’m impressed it actaully scores so well. 2x perf deficiency means M3 Max should catch up easily.